ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-19 22:39:03 +00:00
Code Issues Packages Projects Releases Wiki Activity
Files
440358c16851de74575798c539feca1b0be0799f
composable_kernel/include/ck/tensor_operation/gpu/grid
History
Enrico Degregori 440358c168 Wave Tile Transfer supporting global load with transpose (#3027) 
* Initial implementation:

 - add new thread group transfer supporting transpose instruction
 - refactor AB transfer to switch between thread and wave tiles methods

* Add some comments and remove explicit wave and lane calculations

* Remove compiler option for performance

* fp16 example: use tuned instance

* Missing cleanup

* Integrate wave transfer in existing gemm and batched gemm instances

* Add fast instances

* extend implementation for 8 bit datatypes

packed types not supported

* Address review comments

* Optimize pipeline v1 and re-introduce compiler option

* Disable wave tile approach for b scale gemm

* Fix for clang20

* Avoid code duplication of amd_global_load_transpose_to_vgpr function
2025-10-16 11:33:56 -07:00
..
batchnorm_multiblock
Do not use warpSize as compile time constant as it is removed (#2320)
2025-06-17 11:54:30 -07:00
gemm_layernorm
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
normalization
upgrade from clang-format-12 to clang-format-18 (#2568)
2025-07-28 11:34:07 -07:00
block_to_ctile_map.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_2d_multiple_reduction_multiblock.hpp
Fix for beta!=0 in reduce (#1440)
2024-08-06 09:10:39 -07:00
gridwise_2d_multiple_reduction_threadwise.hpp
update copyright headers (#726)
2023-05-31 18:46:57 -05:00
gridwise_2d_reduction_multiblock.hpp
Fix for beta!=0 in reduce (#1440)
2024-08-06 09:10:39 -07:00
gridwise_2d_reduction_threadwise_multi_d.hpp
upgrade from clang-format-12 to clang-format-18 (#2568)
2025-07-28 11:34:07 -07:00
gridwise_2d_reduction_threadwise.hpp
update copyright headers (#726)
2023-05-31 18:46:57 -05:00
gridwise_ab_transfer_thread_tiles.hpp
Wave Tile Transfer supporting global load with transpose (#3027)
2025-10-16 11:33:56 -07:00
gridwise_ab_transfer_wave_tiles.hpp
Wave Tile Transfer supporting global load with transpose (#3027)
2025-10-16 11:33:56 -07:00
gridwise_batched_gemm_gemm_wmma_cshuffle_v3.hpp
Disable GridwiseOp prints if env var is off (#2843)
2025-09-16 17:47:28 +02:00
gridwise_batched_gemm_gemm_xdl_cshuffle_v1.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_batched_gemm_multiple_d_gemm_multiple_d_xdl_cshuffle_v1.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_batched_gemm_multiple_d_softmax_gemm_xdl_cshuffle_v1.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_batched_gemm_softmax_gemm_wmma_cshuffle.hpp
Disable GridwiseOp prints if env var is off (#2843)
2025-09-16 17:47:28 +02:00
gridwise_batched_gemm_softmax_gemm_xdl_cshuffle_v1.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_batchnorm_backward_blockwise_welford.hpp
update copyright headers (#726)
2023-05-31 18:46:57 -05:00
gridwise_batchnorm_forward_blockwise_welford.hpp
update copyright headers (#726)
2023-05-31 18:46:57 -05:00
gridwise_elementwise_1d_scale.hpp
Refactor elementwise kernels (#1222)
2024-04-19 13:31:17 +02:00
gridwise_elementwise_2d.hpp
upgrade from clang-format-12 to clang-format-18 (#2568)
2025-07-28 11:34:07 -07:00
gridwise_elementwise_layernorm_welford_variance.hpp
enable compilation of INSTANCES_ONLY for Windows (#1082)
2023-12-20 14:34:53 -08:00
gridwise_fpAintB_gemm_wmma.hpp
Disable GridwiseOp prints if env var is off (#2843)
2025-09-16 17:47:28 +02:00
gridwise_gemm_bias_add_reduce_xdl_cshuffle_v1.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_dl_multiple_d.hpp
Add Grouped Conv Fwd Large Tensor kernel (#1432)
2024-08-06 10:06:10 +02:00
gridwise_gemm_dl_v1r3.hpp
upgrade from clang-format-12 to clang-format-18 (#2568)
2025-07-28 11:34:07 -07:00
gridwise_gemm_dpp.hpp
Remove !defined(__HIP_DEVICE_COMPILE__) in CK kernel (#2564)
2025-07-28 13:01:07 -07:00
gridwise_gemm_multiple_abd_xdl_cshuffle.hpp
Conv:TF32: add more instances - 1 (#2867)
2025-09-25 09:27:18 +08:00
gridwise_gemm_multiple_d_multiple_r_xdl_cshuffle.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_multiple_d_wmma_cshuffle.hpp
Disable GridwiseOp prints if env var is off (#2843)
2025-09-16 17:47:28 +02:00
gridwise_gemm_multiple_d_xdl_cshuffle_lds_direct_load.hpp
TF32 POC in Conv3d on MI30x platform #2763 (second attempt) (#2852)
2025-09-17 14:50:15 -07:00
gridwise_gemm_multiple_d_xdl_cshuffle.hpp
[CK] Fix example_grouped_conv_bwd_data_xdl_fp16 with ksplit = 2 (#2943)
2025-09-29 07:56:33 -07:00
gridwise_gemm_multiple_d_xdl_splitk_cshuffle.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_pipeline_selector.hpp
Rebase the PR #1520 to ROCm repo. (#1574)
2025-02-20 18:58:14 -08:00
gridwise_gemm_pipeline_v1.hpp
Navi3 rel (#1176)
2024-03-08 17:11:51 -08:00
gridwise_gemm_pipeline_v2.hpp
Fixed Gemmv2r3 kpad (#938)
2023-09-26 18:40:00 -05:00
gridwise_gemm_pipeline_v3.hpp
initial stream-k implementation with example (#699)
2023-07-26 14:18:15 -05:00
gridwise_gemm_pipeline_v4_direct_load.hpp
Add support for double buffering in direct load GEMM kernel (#1052)
2023-12-03 23:08:47 +01:00
gridwise_gemm_reduce_xdl_cshuffle_v1.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_split_k_multiple_d_xdl_cshuffle_v2.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_split_k_multiple_d_xdl_cshuffle.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_waveletmodel.hpp
update copyright headers (#726)
2023-05-31 18:46:57 -05:00
gridwise_gemm_wmma_cshuffle_v3_b_scale.hpp
Wave Tile Transfer supporting global load with transpose (#3027)
2025-10-16 11:33:56 -07:00
gridwise_gemm_wmma_cshuffle_v3_common.hpp
Wave Tile Transfer supporting global load with transpose (#3027)
2025-10-16 11:33:56 -07:00
gridwise_gemm_wmma_cshuffle_v3.hpp
Wave Tile Transfer supporting global load with transpose (#3027)
2025-10-16 11:33:56 -07:00
gridwise_gemm_wmma.hpp
Disable GridwiseOp prints if env var is off (#2843)
2025-09-16 17:47:28 +02:00
gridwise_gemm_xdl_cshuffle_conv_v3.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_cshuffle_streamk_v3.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_cshuffle_v1.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_cshuffle_v2.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_cshuffle_v3_b_preshuffle.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_cshuffle_v3_b_scale.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_cshuffle_v3_multi_abd.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_cshuffle_v3_multi_d_ab_scale.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_cshuffle_v3_multi_d_b_preshuffle.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_cshuffle_v3_multi_d_blockscale_b_preshuffle.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_cshuffle_v3_multi_d.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_cshuffle_v3_mx_bpreshuffle.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_cshuffle_v3_mx.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_cshuffle_v3.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_layernorm_cshuffle_v1.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdl_waveletmodel_cshuffle.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdlops_bwd_weight.hpp
Conv:TF32: add more instances - 2 (#2879)
2025-10-10 15:28:17 +08:00
gridwise_gemm_xdlops_skip_b_lds_v1.hpp
[CK] Fix misc issues in CK examples (#2890)
2025-09-24 11:28:20 -07:00
gridwise_gemm_xdlops_splitk_lds_direct_load.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdlops_streamk.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdlops_v2r3.hpp
Conv:TF32: add more instances - 1 (#2867)
2025-09-25 09:27:18 +08:00
gridwise_gemm_xdlops_v2r4.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdlops_v2r4r2.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdlops_v3r1.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdlops_v3r2.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_gemm_xdlops_v3r3.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_moe_gemm_blockscale.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_moe_gemm.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_moe_mx_gemm_bns.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_moe_mx_gemm_bpreshuffle.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_moe_mx_gemm.hpp
Extend XDL kernel to Support RDNA3/4 - Part 3 (#2723)
2025-09-09 11:22:36 +08:00
gridwise_permute.hpp
upgrade from clang-format-12 to clang-format-18 (#2568)
2025-07-28 11:34:07 -07:00
gridwise_put_element_1d.hpp
Average pool backward deviceOP and example (#797)
2023-08-10 12:04:35 +08:00
gridwise_set_buffer_value.hpp
update copyright headers (#726)
2023-05-31 18:46:57 -05:00
gridwise_set_multiple_buffer_value.hpp
update copyright headers (#726)
2023-05-31 18:46:57 -05:00
gridwise_softmax.hpp
update copyright headers (#726)
2023-05-31 18:46:57 -05:00
gridwise_sparse_embeddings_forward_layernorm_builtins.hpp
Replace buffer load/store intrinsics with builtins (#1876)
2025-03-05 14:33:28 -08:00
gridwise_sparse_embeddings_forward_layernorm.hpp
[HotFix] add config and version files to pass on build info (#856)
2023-08-23 11:36:17 -07:00
gridwise_tensor_rearrange.hpp
Remove !defined(__HIP_DEVICE_COMPILE__) in CK kernel (#2564)
2025-07-28 13:01:07 -07:00